This article is more than 5 years old. Information may no longer be current.
AVOID-HF: Adjustable ultrafiltration reduces events after HF hospitalization
NATIONAL HARBOR, Md. — Patients hospitalized for acute decompensated HF who were treated with adjustable ultrafiltration had a longer time to first HF events and fewer HF- and CV-related events overall compared with patients who were treated with adjustable loop diuretics, according to results of the AVOID-HF trial presented at the Heart Failure Society of America Annual Scientific Meeting.
However, ultrafiltration was associated with an increase in certain adverse events.
The multicenter, prospective, unblinded trial was designed to include a population of 810 patients with a primary diagnosis of decompensated HF and two or more indicators of fluid overload. However, the trial was stopped early in April 2014 by the sponsor (Baxter Healthcare) after enrollment of 224 patients (27.5%).
Within 24 hours of hospitalization, patients were randomly assigned to adjustable ultrafiltration using the Aquadex FlexFlow System (Baxter Healthcare; n = 110) or adjustable IV loop diuretics (n = 114). The primary endpoint was time to first HF event within 90 days of discharge.
At 90 days, HF events occurred in 25% of patients in the ultrafiltration group vs. 35% of the loop-diuretics group. The estimated time to first HF event was 62 days in the ultrafiltration group vs. 34 days in the loop-diuretics group (P = .106). During a presentation here, Maria Rosa Costanzo, MD, medical director of the Edward Hospital Center for Advanced Heart Failure, Naperville, Ill., attributed the lack of statistical significance to the smaller-than-anticipated sample size.
Patients treated with ultrafiltration were significantly less likely to experience HF-related rehospitalization (P = .034) or CV-related rehospitalization (P = .042) at 30 days, and spent significantly fewer days in the hospital due to readmissions for either cause.
The rate of mortality at 90 days was 15% in the ultrafiltration group vs. 13% in the loop-diuretics group (P = .827). No deaths in either group were considered related to treatment, according to the researchers.
Overall, adverse events were similar in the two groups. Adverse events of special interest, including ACS, a hemoglobin decrease > 3 g/dL, symptomatic hypotension requiring intervention, bleeding resulting in transfusion and central-line bloodstream infections, were significantly more common in the ultrafiltration group (P = .018). Serious treatment-related adverse events also occurred more frequently in those assigned ultrafiltration (14.6% vs. 5.4%; P = .026); however, serious adverse events of special interest, defined as ACS, infection or bleeding requiring transfusion, occurred at similar rates between the groups.
Daniel B. Mark
Costanzo noted that the decision to terminate the study early was made without advance review of trial data or prior consultation with the study’s data and safety monitoring board or steering committee, and that the termination was not related to any safety concerns. She added that, while the sponsor attributed the early termination to slow enrollment, the enrollment rates for AVOID-HF were comparable to those for 154 HF trials published between 2001 and 2012. The early termination of the study was “undoubtedly” its principal limitation, and that the results “should be interpreted with great caution” as a result, she said.
“Decongestion with [adjustable ultrafiltration] must balance the benefit of reducing HF events with the risk of adverse effects,” Costanzo said. “I truly believe that additional investigation of this method of fluid removal is sorely needed.”
Christopher M. O'Connor
In a related editorial, Daniel B. Mark, MD, MPH, director of outcomes research at Duke Clinical Research Institute, and Christopher M. O’Connor, MD, CEO and executive director of the Inova Heart and Vascular Institute, Fairfax, Virginia, noted that the data resulting from the early termination of the trial are not sufficiently clear to allow for favorable guideline recommendations for ultrafiltration or to introduce it as the standard of care.
“In our current regulatory environment … our best defense against more AVOID-HF cases is to only start trials that are designed effectively, budgeted adequately and clearly feasible in terms of enrollment targets,” Mark and O’Connor wrote. “If we can achieve that, we will have done much to ensure that the contribution of patients enrolling in our trials is not wasted.” – by Adam Taliercio
References:
Costanzo MR, et al. Late Breaking Clinical Trials. Presented at: Heart Failure Society of America Annual Scientific Meeting; Sept. 26-29, 2015; National Harbor, Md.
Costanzo MR, et al. JCHF. 2015;doi:10.1016/j.jchf.2015.08.005.
Mark DB, O’Connor CM. JCHF. 2015;doi:10.1016/j.jchf.2015.09.001.
Disclosure: Costanzo, Mark and O’Connor report no relevant financial disclosures. Several researchers report receiving compensation for their work for the AVOID-HF trial steering committee. All researchers’ institutions received research grants from Gambro and Baxter Healthcare for conducting the trial. One researcher is a former employee of and another is a former consultant for Baxter Healthcare.
Perspective
Back to Top
This was an extremely important trial that needed to be done. It was going well, it was conducted to a very high standard, and I think it is tragic that it was stopped early. I’m not aware of any important trial like that having been stopped before for what appears to be business reasons. It’s not as if this was a trial that appeared to be in terrible trouble: The recruitment rate was pretty much what the recruitment rate usually is in HF trials. The final results of this smaller-than-planned trial did suggest that there might be a trend in favor of the ultrafiltration treatment over conventional treatment with IV diuretics. So it’s sort of a double tragedy to, first of all, stop the trial prematurely, and secondly, to see what might’ve been a significant result not be significant as a result of low numbers.
I think the results of this trial reawaken interest in eventually doing another trial in a similar population with the same sort of goal. The key difference between that and some of the earlier trials we’ve seen is that the ultrafiltration was titrated according to the patients’ status. We need to get a definitive trial using that approach. We obviously can learn a lot from [this study], but it must make this treatment attractive for a future clinical trial.